The military history of Imperial China before the nineteenth-century Western impact shows considerable variation from period to period, depending on changing historical circumstances and the differing social bases of successive dynasties. It also shows continuity related to the persistence of the major cultural factors that came together in the Han period. These cultural factors include Confucianism, the Legalist state, and hostility to the nomads of Inner Asia. All three of these emerged individually during the Warring States period that preceded the Qin unification, but should be viewed analytically as part of Imperial China.

Military Officers and Soldiers

Over the long run of Imperial China, the military service obligation of the general population evolved from being nearly universal, as in the Qin and Han, to a burden imposed on a minority. While both the Tang fubing system and the Ming weisuo system employed the principle of soldier-farmers liable to conscription, in both dynasties this principle applied only to a minority of the population. In the Tang fubing membership seems to have been seen as a benefit in the early reigns of the dynasty, later evolving into a burden, while in the Ming weisuo membership seems to have been viewed as a burden from the beginning. In the Song the troops of the standing army were poorly paid and used for menial work, while military officer status was conferred on many officials doing low-level work disdained by true scholar-officials. Coupled with the hypertrophy of “civil” values among the educated elite, these attitudes and patterns of treatment led to the denigration of soldiers (including officers), noticeable from Song times onward and expressed in the often-quoted saying, “Good iron isn’t used for nails; good men aren’t used as soldiers.” Occasional efforts of civil officials to revive the militia ideal of classical antiquity seldom worked as intended.

Weapons and Military Technology

China has been an “advanced” country, in comparison to its contemporaries, through most of recorded history, losing ground only after the Industrial Revolution began in Britain. The major innovations in weaponry that influenced Western military history have their counterparts in China. Any list would include the crossbow, armor, the stirrup, fortifications, gunpowder weapons, and shipbuilding.

In the Qin and Han conscript armies, infantry and cavalry replaced chariots as the principal arm, and the infantry were armed with spears, bows, and in particular crossbows (nu), a weapon in whose technology the Chinese remained superior. Later descriptions of Chinese armies usually include units of archers mixed with crossbowmen, the latter presumably needing protection between rounds due to their longer reloading time. The intricate trigger latch mechanism of the crossbow was a closely guarded state secret under the Han. Battle accounts (too often, unfortunately, influenced by literary conventions) often mention the sky being darkened by clouds of arrows. Evidence for the actual conduct of battles is sketchy, but discharges of arrows (including crossbow bolts) were crucial to victory. Even though infantry bearing shields, swords, and spears existed, there is no trace of either a “phalanx” or a “legion” style of infantry fighting.

Qin Shi Huangdi’s tomb army is wearing armor, and there are many later representations of armored Chinese soldiers. Most of the armor is of the lamellar variety, in which overlapping leather or metal plates of varying size are laced together. Such armor is relatively light and flexible at the expense of protective strength, and in the West infantry and cavalry trained for shock tactics and reliance on edged weapons tended to move on to armor composed at least partly of large plates, of which there are a few Chinese examples.

The idea that the stirrup, by permitting the evolution of shock cavalry armed with the lance, was a primary factor in the creation of European feudalism has received a surprising degree of credence, though recent reevalution of the four-horned Roman saddle has undermined its central thesis.2 In China the spread of the stirrup is associated with the development of the armored cavalryman, mounted on an armored (barded) horse and armed with a lance. Literary references to “armored cavalry” occur as late as the Tang, and a vivid pictorial representation of mounted warriors looking like European knights occurs in a tomb dated to 357 c.E. Nevertheless, it may be stated with confidence that the social outcomes attributed to the stirrup in Europe did not occur in China. Knightlike cavalry were part of the ruling class of north China during the Northern and Southern Dynasties period. This class, which evolved into the governing aristocracy of the Sui and Tang, was largely Xianbi in origin but also included other Inner Asian peoples and Chinese who had adopted barbarian ways. Far from devolving into feudalism, the Sui and Tang dynasties erected a powerful and enduring version of the centralized, bureaucratic empire previously built by the Qin and Han. And, stirrupped or not, the cavalry future belonged to the Inner Asian warrior whose strength was his skill with the bow rather than the lance.

China has always been a country of cities rather than castles, and city walls were not only a means of defense but also a symbol of the city’s status in the hierarchy of rule. The walls were formidable defenses. While there are many recorded examples of long sieges and much literature on siege-craft, it remained the case that the best way to take a city was by treachery or surprise during a period of confusion, and a siege was more likely to be won by protracted blockade than by a successful assault. China’s urban fortifications did not evolve the low, relatively cannon-proof bastions of the trace italienne, as European cities did in the sixteenth century while China lived peacefully under the rule of the Ming. Afterward, the thick earthen walls of the major Chinese cities remained highly resistant to the gunpowder weapons that were becoming more prominent in Chinese warfare.

The basic formula for gunpowder was known to the Song, weapons incorporating gunpowder were used prominently during the Yuan, and in the Ming Yongle reign (1402-1424) a special headquarters was established in Beijing to coordinate the training of gunners. Firearms added to the defensive strength of the Great Wall, itself a Ming creation, and the Chinese element of the Manchu banner system seem to have been valued, in part, as artillery specialists. However, we cannot discern a “gunpowder revolution” in Chinese military history. In the Ming, Qi Jiguang’s successful and widely emulated military organization had gunners serving alongside bowmen, swordsmen, and spearmen in the same primary (squad-level) formations, and in the Qing Zeng Guofan’s Hunan Army battalions combined newer and older weapons in the same way. Firearms originated in China, but in China they remained just another missile weapon. One does not see efforts to standardize manufacture, reduce the number of calibers, or create new tactics and organizations to exploit the potential of a new weapons system.

Marco Polo’s descriptions of Chinese ships were part of his credibility problem in Europe, and Europeans also found it difficult to credit the early Ming naval voyages. It is now accepted that China built wooden ships as large or larger than any ever built in Europe, and, having invented the compass, navigated them beyond the sight of land to Africa and other distant coasts. But these capabilities did not add up to a navy; in the latter part of the Ming and in the Qing, China’s seagoing forces consisted of small ships and boats tethered to the military organizations of specific provinces.

China’s long history of technological progress provides scant comfort for theories that see certain kinds of social and political change as the inevitable result of specific technologies. Neither the stirrup nor gunpowder had the dramatic consequences in China claimed for them in Europe. With respect to shipbuilding technology, Ming China’s withdrawal from the sea was deliberate and dramatic, and had long-lasting consequences. It compares to Tokugawa Japan’s “giving up the gun.”3 In both cases, ruling establishments feared and prevented technology-driven change.

Military Institutions

Within the context of the factors of continuity and change already discussed, we may see three broad (and partly overlapping) subperiods in the evolution of Imperial China’s military institutions and practices, each of which transcends any single dynasty, and each of which came to an end due to a crisis of Chinese civilization involving the two basic military threats: domestic rebellion and foreign invasion. The first subperiod is bounded by the rise of Qin in the Warring States and the end of the last of the Six Dynasties in 589 C.E., the second by the consolidation of the Northern Wei in the fifth century and the final Mongol conquest of the Song in 1279, and the third by the Kitan conquest of part of north China in the tenth century and the fall of the imperial system as a whole in the twentieth century. We will label these subperiods Han, Tang, and Mongol-Manchu.


The two Han dynasties continued to employ the cadre-conscript army developed by the state of Qin during the Warring States, just as they continued the bureaucratic system and other Qin institutions. Similarly, the military systems of the Three Kingdoms, the ephemeral Western Jin (265-316), and the later south China regimes collectively called the Six Dynasties evolved from the Later Han state of affairs in which rival warlords controlled armies of dependent soldiers (buqu).

The career and reforms of Shang Yang (d. 338 B.C.E.) in Qin are described in a hostile and caricatured way in the sources, but they converted Qin permanently into the strongest of the seven warring states well over a century before the final Qin conquest of China. Shang Yang abolished hereditary status and created a new set of “titles of nobility” (jue) that could be conferred on any male subject, but only for success in war or agriculture. The population was organized in mutual responsibility groups and governed by officials who could not be natives of the areas they governed. These officials were rewarded (or punished) strictly for their success (or failure) in carrying out their orders. The other states contemporary with Qin undertook less comprehensive and less successful reforms, but Qin retained the leadership that Shang Yang’s reforms had conferred. Qin’s greatest general, Bai Qi (d. 257 B.C.E.), made it a deliberate policy to massacre the armies of the states he defeated in order to maintain Qin’s comparative advantage.

While the first Han emperor made a great show of moderating the severity of Qin laws and experimented with a limited revival of feudalism, in the end the Han continued most Qin institutions, including the Qin military system. For most people conscription was the most important element of that system. Men were drafted for two years, serving as infantry, cavalry, or sailors according to their background. For a small minority this meant service in the capital, and for a larger minority service along the walled defenses of the northern frontier, whose operation in Han times is understood in unusual detail from surviving contemporary documents.4 Most conscripts seem to have served their time within their native province (jun, “commandery”), whose governor (taishou, literally “grand defender”) was also their commander in case of invasion. The founding of the Han coincided closely with the unification of the Xiongnu under Maodun, and Han Wudi’s resort to war against the Xiongnu is associated with the creation of specialist cavalry forces that could fight in the Xiongnu manner, most famously by Huo Qubing (d. 117 B.C.E.). But Wudi’s wars against the Xiongnu and his annexations of territory in Korea, south China, and Vietnam were made possible by the mobilization of large numbers of mostly infantry troops, and this capacity was retained under his successors.

Guangwudi (r. 25-57), the founding emperor of the Later Han, lightened the military burden by eliminating the annual summer mobilization of the reservists. The Later Han maintained military pressure on the Xiongnu, and finally broke them up for good. Except for the adventures of Ban Chao (d. 102) in the Western Regions (now Xinjiang), which were a classic example of indirect rule maintained by locally recruited troops, the Later Han was not committed to territorial expansion. Despite coups and conflicts in Luoyang, relative peace prevailed in the provinces, along with increasing concentration of landownership. When the Later Han confronted its major military crisis, the Yellow Turban rebellion (from 184), the fastest way to mobilize large armies was to recruit among the dependent clients of already powerful notables; a breakdown to war-lordism followed quickly.

Cao Cao (155-220) was the most successful of these warlords, and his descendants were the rulers of Wei, the most powerful of the Three Kingdoms. His rivals founded Shu-Han (221-263, in Sichuan) and Wu (formally 229-280, at Nanjing). The Jin dynasty of Sima Yi and his descendants ended the Three Kingdoms and briefly ruled over a reunified China. After the rebellions and invasions of the early fourth century, the Jin ruled south China from Nanjing until 420, where four more Chinese dynasties followed until 589.

Many scholars believe that under these dynasties peasants were reduced to the status of serfs, and that armies also were composed of soldiers who were unfree dependents (buqu). While some of this theorizing is in the service of a Marxist periodization of Chinese history, it is very clear from the histories of these dynasties that a warlord pattern had developed: For whatever reason, soldiers were at the disposal of their generals, and central authority was correspondingly fragile. While expressions of disdain for soldiers can be found in the literature of the period, many eminent literary figures also exercised high military command, and the warlord founders of two dynasties (Liu-Song and Liang) had sons who compiled major literary collections (Liu Yiqing and Xiao Tong, compilers of the Shishuo xinyu and the Wenx-uan, respectively). The Sui conquest of Nanjing ended this line of evolution.


In 493 Tuoba Hong, the Northern Wei emperor posthumously titled Xiaowendi, played a trick on his Xianbi clan leaders. Pretending to lead them in an invasion of south China, he instead made them stop at the still impressive ruins of Luoyang, the capital of the Later Han and Western Jin, which he made his own capital. North China had been overrun early in the fourth century by various Inner Asian peoples who diplayed an uncharacteristic hostility to Chinese civilization. After the disorders of this period, the brief stabilization of the Northern Wei in the fifth century as the first of the important “dynasties of conquest” begins the second period of military evolution. The Northern Wei created early forms of the equal field (juntian) land system and the fubing military system that became major institutions under the Sui and Tang dynasties. Most important, the Northern Wei attempted to create a society in which the military skills of the Xianbi would be complemented by bureaucratic and literary skills of the Chinese educated elite. Later dynasties of conquest made the same attempt, and in military matters Inner Asian influence was important even in dynasties (Sui, Tang, Ming) usually considered Chinese.

After the breakup of the Northern Wei, Yuwen Tai (505-556) and his descendants ruled the northwest first through puppet emperors of Western Wei and then as emperors of the Northern Zhou, and there both the soldier-farmer (fubing) military system and the mixed Chinese and Inner Asian Guanzhong aristocracy that commanded it evolved to provide military means and leaderhip for the Sui and Tang empires. The Yuwen rulers were not of Chinese origin, while the Sui founder and the father of the Tang founder were married to sisters from the Xiongnu Dugu clan. By the end of the sixth century, surnames within the Guanzhong aristocracy did not indicate purely Chinese or Inner Asian ancestry because of intermarriage, and similarly the fubing soldiers included elements capable of fighting on foot or on horseback. Under the fubing system each headquarters (fu) commanded about one thousand farmer-soldiers who could be mobilized for war. In peacetime they were self-sustaining on their land allotments, and were obliged to do tours of active duty in the capital. These tours were usually one month long (two months for the most distant units), and their frequency depended on the distance of each unit from the capital. The fubing soldiers permitted the Sui and Tang founders to conquer China, but attempts at foreign conquest were less consistently successful. Obsessive efforts to subdue the Korean kingdom of Koguryo ultimately cost the second Sui emperor his throne and his life. Tang Taizong (r. 626-649) fought both Türks and Tibetans to peace on favorable terms, but failed to overcome Koguryo. That goal was accomplished by his son Gaozong (r. 649-683), though the final winner was not Tang China but its ally, the southern Korean kingdom of Silla, which succeeded in unifying the entire peninsula under its own rule. Japan, which had supported Paekche, the third Korean kingdom, was alarmed by these developments and responded by imitating the fubing and other Tang institutions in the Taika reforms.

Most of the fubing units were located in the northwest, and the system was best suited for the annual campaigning cycle of an expanding empire. Under Empress Wu (r. 684-705) the fubing system declined, and under Xuanzong (r. 712-756) a standing army stationed on the northern frontier evolved in its place. This army reached a strength of half a million men and eighty thousand horses by the 740s. Its Chinese personnel included many men displaced by economic changes since the founding of the Tang, and its non-Chinese personnel included Koreans, Kitan, Türks, and Sogdians. The new standing army thus preserved the Chinese-Inner Asian mixture characteristic of the early Tang, but the old Guanzhong aristocracy ceased to have much involvement with it and its higher ranks came to be filled from within. Having accepted the decline to uselessness of the fubing system, the Tang court had no central army to resist the An Lushan rebellion, and could only counter it by appealing to other frontier commanders whose social background was similar to An Lushan’s and who could move swiftly from loyalty to rebellion when their autonomy was challenged. Despite impressive successes by the court, the pattern of regional warlordism continued until the fall of Tang. While the replacement of the fubing system with the standing army was a major discontinuity in China’s military development, this discontinuity occurred in a period of peace as a result of a deliberate policy decision of the Tang government. While it led to disorder, it was not caused by defeat.

Recognizing the need for a central army as a counterweight to the troops of the regional warlords, the post-An Lushan Tang emperors created the Divine Strategy (Shence) Armies, whose eunuch commanders grew increasingly powerful as the Tang declined. The Privy Council or Bureau of Military Affairs (Shumiyuan), originally a eunuch agency, was taken over by generals during the Five Dynasties (907-960), while continuing to command the central armies (jinjun, qinjun) at the personal disposal of the emperors. The Five Dynasties were politically unstable, each ending in a violent overthrow, but they were militarily successful, since the territory ruled from Luoyang expanded and the troops were increasingly concentrated in the central armies.

The Song founder continued this system, making modifications in the interest of political stability. He retired his principal generals, turned the Bureau of Military Affairs into a department controlled by civil officials, and moved the capital to Kaifeng to make supply via the Grand Canal easier. The chain of command over the central army troops concentrated in the capital area was changed regularly to prevent any general from developing a dangerous personal ascendancy over a particular body of troops. Under the first three Song emperors, the army was efficient enough to reunify the south Chinese states (the Ten Kingdoms) with the empire, but was not strong enough to destroy the two states ruled by Inner Asian peoples (Tangut Xixia and Kitan Liao) that together dominated the northern frontier. The long-term trend in the Northern Song was for the central army to become larger and more expensive, while its soldiers became poorer and less capable militarily and its civilian administrators more intrusive and abusive. The relative ease with which the Jurchen Jin conquered Kaifeng and the rest of north China illustrated the decay to the Song military system. The Hangzhou-based Southern Song depended militarily on an exiguous combination of warlord-led improvised armies and naval power (exercised along the Yangzi as well as on the ocean). The execution of Yue Fei, the most prominent of the warlords, restored political stability even as it dimmed the hope of reconquering the north. When the Mongols completed the destruction of the Southern Song in the 1270s, they ended both the much-discussed “early modern” economic developments of the Song and the continuous line of military evolution that had begun in the Northern Wei.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s